Contents lists available at ScienceDirect

Tectonophysics

journal homepage: www.elsevier.com/locate/tecto

New paleomagnetic pole for the Upper Jurassic Chon Aike Formation ofsouthern Argentina (South America): Testing the tectonic stability ofPatagonia with respect to South America, and implications to MiddleJurassic-Early Cretaceous true polar wanderVíctor Ruiz Gonzáleza,⁎, Carla G. Puigdomenecha, Emiliano M. Rendaa, Bárbara Boltshausera,Rubén Somozaa, Haroldo Vizána, Claudia B. Zaffaranab, Graeme K. Taylorc, Miguel Hallerd,Raúl Fernándeze

a IGEBA, UBA – CONICET, Argentinab IIPG, UNRN – CONICET, Argentinac Plymouth University, United Kingdomd CENPAT, CONICET – UNPSJB, Argentinae INREMI, UNLP, Argentina

A R T I C L E I N F O

Keywords:Chon Aike FormationDeseado MassifSouth AmericaPaleomagnetismApparent polar wander path

A B S T R A C T

A paleomagnetic pole and a new Ar-Ar date of 156.08 ± 0.05 Ma are reported for the Chon Aike Formation, anextensive plateau of ignimbrites, outcropping in the Deseado Massif (southern Patagonia, Argentina). Thegeographic coordinates and statistical parameters for the paleomagnetic pole, validated by a reversal test, are:Lat. = 84.3° S, Long. = 191.3° E, A95 = 8.6°, K = 13.3, N = 23. This pole shows a complete coherence with LateJurassic and Early to mid-Cretaceous poles of South America, obtained exclusively from igneous rocks, avoidingany potential inclination shallowing. It is suggested that; a) southern Patagonia was an integral part of SouthAmerica in the Late Jurassic; b) any reported vertical axis tectonic rotations must pre-date the Late Jurassic inthis region; c) according to updated paleomagnetic data from igneous rocks, South America had very limitedlatitudinal drift between about 160 and 140 Ma; and, therefore, d) its apparent polar wander (APW) path, whencompared with global apparent polar wander paths, does not support a large global shift or true polar wander(TPW) between 160 and 140 Ma.

1. Introduction

Paleomagnetism is an important tool for paleogeographic re-constructions and geodynamic analysis. This can be achieved by com-parison of apparent polar wander (APW) paths which are a temporalsequence of paleomagnetic poles, obtained from a variety of rock types.APW paths record the tectonic movements of lithospheric plates and, ifactive, true polar wander (TPW) which is the displacement of an entire,rigid Earth with respect to its spin axis (Goldreich and Toomre, 1969;Gordon and Livermore, 1987). Paleomagnetic poles are also very usefulto test tectonic rotations (around a vertical axis) of lithospheric blocksdue to the activity of fault systems with lateral displacements.

Using an APW track of paleomagnetic poles, Kent et al. (2015) havesuggested that a TPW episode took place between about 160 and

140 Ma, which should be testable by reliable paleomagnetic poles fromwell-dated units of different plates around the globe.

The aim of the present work is to constrain the timing of blocktectonic rotations in the Deseado Massif of southern Patagonia and totest if paleomagnetic data from South America supports the proposal ofa large shift of the whole rigid Earth during the Late Jurassic. To fulfillthese goals, paleomagnetic sampling was carried out in the Chon AikeFormation. Its outcrops in the Deseado Massif (Santa Cruz Province -Argentina) are considered an integral part of South American Plateduring Jurassic times (von Gosen and Loske, 2004; Zaffarana et al.,2010).

A reliable paleomagnetic pole was obtained from several volcanicflows for which geographic coordinates and statistical parameters are:Lat. = 84.3° S, Long. = 191.3° E, A95 = 8.6°, K = 13.3, N = 23. The

https://doi.org/10.1016/j.tecto.2018.10.028Received 5 January 2018; Received in revised form 21 October 2018; Accepted 26 October 2018

⁎ Corresponding author at: Instituto de Geociencias Básicas, Aplicadas y Ambientales de Buenos Aires, UBA – CONICET, Oficina 27 - Departamento de CienciasGeológicas, Facultad de Ciencias Exactas y Naturales – UBA, Intendente Güiraldes 2160, CABA, Argentina.

E-mail address: vruizgonzalez@gl.fcen.uba.ar (V. Ruiz González).

Tectonophysics 750 (2019) 45–55

Available online 15 November 20180040-1951/ © 2018 Elsevier B.V. All rights reserved.

T

Chon Aike Formation's paleomagnetic sites belong to a voluminous unitof ignimbrites, in which fiammes indicate that it was deposited hor-izontal to sub-horizontally and an absence of subsequent deformation.Only one sampling site (number 30 in Table 1) belongs to a bed with adip of 65° that was later discarded given the level of uncertainty of itspaleomagnetic direction (see below). The northern part of the forma-tion has been well-constrained age-wise (Pankhurst et al., 2000; Féraudet al., 1999), however, its outcrops cover a large area and there is onlyone available age in its central area and two in the southern part (seeFig. 1).

The new pole for the Chon Aike Formation, presented in this work,precisely constrains the age of tectonic rotations to the Late Jurassic,possibly before the breakup of Western Gondwana (see Somoza et al.,2008). Moreover, when it is compared with previously reported pa-leomagnetic poles from Early Jurassic to mid-Cretaceous igneous unitsof South America, it precludes any appreciable latitudinal drift of thiscontinent during the Late Jurassic.

2. Geological setting and paleomagnetic sampling

During Jurassic times, voluminous volcanism developed over a widearea of Patagonia, which was named by Kay et al. (1989) as the ChonAike Magmatic Province. The distribution of this magmatic provinceincludes the North Patagonian and Deseado Massifs (Río Negro, Chubut

and Santa Cruz Argentinian provinces), as well as parts of Tierra delFuego, Antarctica and the South Atlantic Islands Province (Pankhurstet al., 2000). In the extra Andean area of the Deseado Massif, there is animportant sequence of volcanogenic rocks that forms part of thismagmatic province, including the pyro- and epiclastitic lithologies ofthe Roca Blanca Formation, lavas and subvolcanic bodies of the BajoPobre Formation, and the pyro-epiclastitic rocks and lavas of BahíaLaura Group, which includes the Chon Aike Formation (Panza andHaller, 2002).

Paleomagnetic sampling was carried out in an ignimbritic plateau ofabout 18,000 km2, in the western part of the Chon Aike Formation(Fig. 1). This plateau constitutes a sequence of ca. 1,300 m of fallouttuffs and ignimbrites that correspond to several cycles of eruptive ac-tivity (Echeveste et al., 1999). The thickness of each ignimbrite flowvaries between 5 and 15 m (Fig. 2), and, in extreme cases, up to 35 m.Its dominant color in outcrop is dark brown to reddish, sometimes lightand dark gray, while in fresh faces they can show a pinkish to reddishgray and purple coloration. The ignimbrites contain abundant fiammeswith diameters ranging between 2 and ca. 10 cm. Some of thesefiammes have been replaced by calcite-filled cavities. It is important tonote that both the fiammes and the cavities are parallel to subparallel tothe ignimbrite beds, which are horizontal to sub-horizontal. This bed-ding plane is also shown in many cases by biotite sheets within theignimbrites (Fig. 3).

Table 1Site: number of paleomagnetic sampling sites; Lat. And Long.: geographic coordinates of these sites; N/n: number of collected samples vs. Number of directions usedfor the site means; Dec. And Inc.: declination and inclination of the site mean directions; α95: 95% cone of confidence of site mean directions; k: kappa, precisionpoaramenter of Fisher (1953); Pole Lat. And Pole Long.: geographic coordinates of virtual geomagnetic poles (VGPs). * Discarded because of its large confidence cone(α95 = 37.4°), the site also could correspond to an edge of a volcano caldera. ** Sites with VGPs away from the cutt of angle determined using Wilson et al. (1972)method applied iteratively. Calculated mean directions for 30 and 23 sites used to calculate the reversal test (see text).

Site Lat. (°S) Long. (°E) N/n In situ Bedding correction Tilt corrected α95 k VGPs

Dec. Inc. Strike Dip Dec. Inc. Pole Lat. Pole Long.

1 47.5800 −69.7710 8/7 201.4 70.8 Horizontal 201.4 70.8 3.1 373.4 −74.7 237.62 47.6122 −69.7689 6/6 173.1 67.3 Horizontal 173.1 67.3 5.6 146.6 −84.8 359.53 47.6315 −69.7555 5/5 168.2 76.4 Horizontal 168.2 76.4 3.4 516.7 −72.3 307.34 47.7587 −69.7041 5/5 167.9 58.8 Horizontal 167.9 58.8 7.2 112.9 −78.0 59.05 47.8100 −69.7008 11/9 162.1 65.5 Horizontal 162.1 65.5 4.2 153.7 −78.0 14.46** 47.7653 −69.7008 7/7 322.2 85.6 200 10 299.8 76.1 6.6 85.0 −31.0 263.67 47.8145 −69.9409 6/6 9.4 −50.0 Horizontal 9.4 −50.0 4.2 259.5 71.5 316.38 47.8193 −69.9309 7/7 163.1 52.6 Horizontal 163.1 52.6 4.2 207.7 −70.7 62.99 47.9117 −70.0208 5/5 217.5 43.9 Horizontal 217.5 43.9 11.9 42.6 −53.3 176.310 47.9228 −69.8090 7/6 322.0 −48.4 Horizontal 322.0 −48.4 9.8 47.7 55.5 218.711** 47.9266 −69.6652 5/4 288.6 −64.7 Horizontal 288.6 −64.7 6.8 184.9 43.4 173.812 47.9166 −69.6736 6/3 177.2 48.6 Horizontal 177.2 48.6 13.2 88.0 −71.5 102.713 47.9259 −69.5375 4/4 192.0 70.9 Horizontal 192.0 70.9 6.0 235.4 −79.5 249.714** 47.9249 −69.5410 6/6 89.4 18.8 Horizontal 89.4 18.8 6.1 120.2 −6.8 13.515** 47.9336 −69.3577 5/5 157.3 26.7 Horizontal 157.3 26.7 2.6 854.5 −51.3 73.916 47.9297 −69.3687 5/5 240.7 71.2 Horizontal 240.7 71.2 2.7 804.8 −52.9 236.217 46.9686 −70.5638 7/6 187.6 65.0 Horizontal 187.6 65.0 5.5 151.0 −84.8 202.418 47.0178 −70.5098 7/7 342.0 −72.4 Horizontal 342.0 −72.4 3.2 365.5 74.9 148.819 46.9023 −70.4732 9/7 23.1 −75.9 Horizontal 23.1 −75.9 11.2 29.9 69.1 80.020** 47.0215 −70.5414 7/7 108.2 46.6 Horizontal 108.2 46.6 6.9 82.7 −32.0 11.621 46.9457 −70.4842 11/11 171.4 60.2 Horizontal 171.4 60.2 4.0 131.0 −81.5 60.022 47.2290 −70.7265 8/4 9.9 −65.0 Horizontal 9.9 −65.0 18.3 26.1 83.3 21.223 47.2363 −70.5457 8/8 334.8 −69.7 180 10 3.1 −71.6 4.2 175.2 80.6 98.824** 47.4326 −69.5457 9/7 81.6 59.4 Horizontal 81.6 59.4 10.7 33.0 −23.6 345.025 47.3211 −68.5457 7/7 207.2 52.6 Horizontal 207.2 52.6 4.6 170.6 −65.0 174.726 47.3048 −67.5457 6/6 200.7 27.7 Horizontal 200.7 27.7 4.5 223.1 −53.1 144.227** 47.0091 −66.5457 5/5 282.6 74.2 220 5 288.9 69.7 5.1 223.7 −27.1 250.828 47.0693 −65.5457 5/4 32.2 −40.7 Horizontal 32.2 −40.7 5.7 265.0 54.9 348.629 47.1902 −64.5457 5/3 232.2 79.9 200 5 250.3 76.5 11.2 122.4 −49.5 251.330* 47.3898 −63.5457 5/3 136.0 −27.3 225 65 136.1 37.7 37.4 11.9 −58.3 40.331 47.0897 −62.5457 5/3 16.3 −60.6 Horizontal 16.3 −60.6 21.1 35.1 77.1 0.6

Mean directions (paleohorizontal) N Dec. Inc. R k α95

23 sites Reverse 15.0 190.2 62.7 14.3 21.0 8.5Normal 8.0 5.4 −62.6 7.7 21.5 12.2

V. Ruiz González et al. Tectonophysics 750 (2019) 45–55

46

Paleomagnetic sampling was carried out in the northern and centralareas of the plateau (see Fig. 1). In order to average out paleosecularvariation of the Earth magnetic field (EMF), 31 paleomagnetic siteswere sampled from different ignimbrite flows. These samples werecollected using a portable core-drilling machine and oriented usingmagnetic and solar compasses. Five to eleven samples were taken fromeach site (Table 1). The sampling was carried out extending over sev-eral meters laterally along each ignimbrite flow, and high ground wasavoided in order to reduce the possibility of remagnetization of theflows due to lightning strikes.

3. Petrographic description

Petrographic analyses were carried out to characterize their primarymineralogical composition and to determine any post-depositional al-teration that could have affected the original magnetic carriers. Theignimbrites are composed mainly of crystalloclasts of quartz, plagio-clase and biotite or amphibole. Euhedral to subhedral zircons andapatites were detected as minor components. Most of the quartz crystalsshow embayments of corrosion. Plagioclases are zoned (rarely alteredto sericite), and biotites show different degrees of alteration to opaque

Fig. 1. Geologic map of the sampled ignimbritic plateau and its location map. Dated localities are marked with stars and paleomagnetic sampling sites with yellowcircles. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 2. Picture of one of the sequences of ignimbrites in the Río Pinturas creek. Note the thickness of each ignimbrite and the original horizontality.

V. Ruiz González et al. Tectonophysics 750 (2019) 45–55

47

minerals. The matrix of the ignimbrites is recrystallized, including glassshards and showing different degrees of welding. In all cases, opaqueminerals are present, and sometimes rutile appears as an accessory. Thefiammes are devitrified and show strong orientation with evidence offlux (Fig. 4a). Reflected light analysis shows magnetite altered to he-matite that could be the potential magnetic carrier (Fig. 4b).

4. Radiometric dating

A paleomagnetic pole named Chon Aike (CA) has been listed invarious databases (e.g., Torsvik et al., 2012; Kent and Irving, 2010) andhas been assigned an age of 165–170 Ma. However, the units sampledfor that paleomagnetic pole belongs to the broad “Chon Aike MagmaticProvince”, which is composed of several different formations with agesranging between 158.4 Ma (Féraud et al., 1999) and 178.4 Ma(Pankhurst et al., 2000). This huge magmatic province is located farthereast where it is extremely difficult to determine sequences with severalstratigraphic levels of ignimbrites (Somoza et al., 2008). Hence, we are

unclear as to whether the previous pole contains enough individualflows to adequately average out secular variation.

The Chon Aike Formation, which constitutes the ignimbritic plateaustudied here, has been investigated age-wise by various authors usingdifferent dating techniques. In this plateau, there are 6 similar radio-metric age determinations with a mean value of ~154 Ma in thenorthern part (see Fig. 1) and 2 radiometric ages in the southern partwith an average age of 158 Ma. For the rocks sampled in the centralarea, there is only one published age of 152 ± 2.8 Ma by U-Pb SHRIMPin zircons, obtained on a brownish strongly welded ignimbrite (Moreiraet al., 2006).

In this paper, we collected a grayish ignimbrite at 47.96°S, 69.37°W,which underlies and therefore pre-dates the 152 Ma flow. This samplewas dated by the Ar39/Ar40 method in individual biotite crystals(Fig. 5a) and its weighted mean age (from 19 crystals) is156.08 ± 0.05 (2 σ), and therefore similar to the age of the overlyingignimbrite. The sample gave a plateau age of 156.15 ± 0.38 for the39Ar released. There is no statistical difference between the plateau ageand the inverse isochron age (156.82 ± 0.77), and the initial ratio issmaller than atmospheric, demonstrating that the age has not beeninfluenced by excess argon (Fig. 5b). The study of this sample wasundertaken by the OSU Argon Geochronology Lab, CEOAS Oregon StateUniversity, Corvallis. The new radiometric age underpins the age of ourpaleomagnetic sampling sites in the central part of the plateau.

This result confirms that all the ages obtained in the different lo-calities of the ignimbritic plateau range between 151.5 and 158.4 Ma,so the age of our reported paleomagnetic data represents a portion ofthis time interval. We consider it appropriate to assign an average ageof 155.0 ± 3.5 Ma to the Chon Aike Formation paleomagnetic pole(where the error bar corresponds to the range of all reported ages,Fig. 1).

5. Paleomagnetic results

The oriented paleomagnetic samples were processed in theLaboratory of Paleomagnetism “Daniel A. Valencio” in the Departmentof Geological Sciences of the University of Buenos Aires (Argentina).The samples were sliced into one or two specimens of fresh rock, of2.5 cm in diameter and 2.2 cm length. Natural remanent magnetization

Fig. 3. Flux of ignimbrite parallel to bedding planes marked by fiammes andthe fluidal distribution of the minerals.

Fig. 4. Thin section of the sample taken from the site 29. a) fiamme showing fluidity and crystals of plagioclase (Plag), quartz (Qz) and opaque mineral (magnetite?,Mt) in a devitrified matrix; b) crystals of magnetite altered to hematites (left) and rutile with hematite (right).

V. Ruiz González et al. Tectonophysics 750 (2019) 45–55

48

(NRM) before and after demagnetization were determined using anAGICO JR6 magnetometer and a 2 G cryogenic magnetometer whichincludes a three-axis AF demagnetizer. The stability of the NRM wastested by progressive thermal or alternating field (AF) demagnetization.Thermal demagnetization was performed using an ASC scientific oven.Bulk susceptibility measurements were carried out with an MS2Bartington instrument after each heating step to monitor the occurrenceof chemical possible changes induced by heating. Anisotropy of mag-netic susceptibility (AMS) and thermomagnetic curves were determinedwith a multifunction Kappabridge MFK1. The ChRM of each specimenwas determined by applying a least squares line fit (Kirschvink, 1980).

AF demagnetization defined a component of high coercivity after10 mT (Fig. 6a). These samples have thermomagnetic curves with Curietemperatures similar to those of (titano-)magnetites (Fig. 6b). It is ob-served that the spectrum appears to be modified with heating, reachinglower Curie temperatures on cooling, which may be due to the ex-solution of members of the solid solution series. An irreversible drop insusceptibility around 300–400 °C may indicate maghemitization of thesampled rocks (in agreement with reflected light analysis). In few cases,

the use of AF was not effective in completely demagnetizing the sam-ples (Fig. 6c) and IRM studies of these samples show, the presence of ahigh coercivity phase along with (titano)magnetites (Fig. 6d). The IRMacquisition curves of these types of samples shows a sharp rise ofmagnetization in magnetic fields (H) at < 200 mT, followed by a moregradual acquisition of additional IRM in stronger H. The application ofa backfield shows a minimum remanent coercive force (Hcr) around110 mT. For these cases, the thermal demagnetization method was usedexclusively to identify remanent components (Fig. 6e). For the majorityof samples, in specimens of the same sample, both AF and thermaldemagnetizations yielded similar ChRM components (Fig. 6f). In con-sequence, it was possible to define a component of high coercivity andhigh unblocking temperature, showing positive and negative inclina-tions (dual polarity, see Fig. 6a and e). Characteristic remanent mag-netizations (ChRM) where determined with maximum angle deviations(MAD) ≤10°.

Site mean directions were found from between three to elevenChRMs (Fisher, 1953; see Table 1). 30 site mean directions were ac-cepted for the rest of the analysis, following the criteria of Lanza and

Fig. 5. Results of the Ar-Ar method in biotite crystals showing the mean age for the site.

V. Ruiz González et al. Tectonophysics 750 (2019) 45–55

49

Meloni (2006), while one site (30; Table 1) mean direction was dis-carded because of its large confidence cone (α95 = 37.4). Incidentally,the discarded site is the only one that has a bedding dip of 65° and couldcorrespond to an edge of a volcano caldera. In the other 21 (9) sitespositive (negative) inclinations were observed, implying a dual polarityrecord. These data, however, show a high degree of between-site scatterin their mean directions (Fig. 7a) and so anisotropy of magnetic sus-ceptibility (AMS) studies were performed to test if the flow-inducedignimbrite fabric had deflected the ChRM's. Sites 14 and 22, whichshow markedly different ChRM's (see Table 1), display consistent flatlying fabrics (Fig. 8). The mean tensors of site 14 (N = 7) areK1 = 1.008, K2 = 1.005, K3 = 0.987; and those of site 22 (N = 9) are:K1 = 1.007, K2 = 1.004, K3 = 0.989. Together, these data suggest thatthe CHRMs are not affected by the anisotropy, but the scatter in di-rections is a true record of the Earth's Magnetic Field. This proposal isconsistent with Lanza and Zanella (1993), Tarling et al. (1999), Vizánand Van Zele (2001), Mena et al. (2006) who have suggested that theremarkable dispersion of Jurassic-early Cretaceous paleomagnetic datamay be due to larger magnitudes of the secular variation than at

present, or to transitional phases of the geomagnetic field during thistime span of high frequency of reversals (as documented by severalmagnetostratigraphies). It is important to note that in the above-men-tioned works the authors analyzed paleomagnetic data from limestones,dolerites and volcanic rocks, which are not affected by inclinationflattening.

5.1. Validation of the results

Virtual Geomagnetic Poles (VGPs) were calculated from the 30mean site directions, which are dispersed due to geomagnetic fieldbehavior (see Lanza and Zanella, 1993; Tarling et al., 1999; Vizán andVan Zele, 2001; Mena et al., 2006). To determine the Chon Aike For-mation paleomagnetic pole, an iterative method looking for the highestprecision parameter Kappa (Fisher, 1953) in the calculation of theaverage pole was applied: only those VGPs that were inside a cut-offangle of 40° to the mean pole, were averaged (Wilson et al., 1972).Seven VGPs were excluded because they differ > 40° (sites with ** inTable 1, Fig. 9) from the mean pole. Its geographic coordinates and

Fig. 6. a) Zijderveld diagram showing the ChRM and module of a sample demagnetized by AF, with reversal polarity; b) demagnetization high-temperature curvewith Curie temperatures of (titano)magnetites and suggesting those as the main carriers of the remanence; c) AF demagnetization curve showing the presence of ahigh-coercivity mineral; d) IRM and backfield diagram indicating the existence of hematite, e) Zijderveld diagram showing the ChRM and module of a sampledemagnetized using thermal method, with normal polarity but similar direction as in a); f) Zijderveld diagrams of the same sample demagnetized by thermal and AF,showing similar ChRM; g) Zijderveld diagrams of different sample sites with the opposite polarity to that in f) demagnetized by thermal and AF, showing similarChRM.

V. Ruiz González et al. Tectonophysics 750 (2019) 45–55

50

statistical parameters are: Lat. = 84.3° S, Long. = 191.3° E, A95 = 8.6°,K = 13.3, N = 23.

In order to assess the reliability of the calculated pole a reversal testwas carried out (see Table 1), on the associated directions (McFaddenand McElhinny, 1990) using the implementation in the IAPD softwareof Torsvik et al. (2014). Applying the test to the 23 sites that are used inthe calculation of the pole, yields a “Class C” results, with an observedangle of 2.21°, within the critical angle of 14.23° (McFadden andMcElhinny, 1990).

As pointed out by Deenen et al. (2011), among others, secularvariation produces a circular distribution of poles about a mean pole,rather than a circular distribution of directions about a mean direction.It is therefore, perhaps, more appropriate to carry out the reversal teston the distribution of poles. For this purpose, the paleomagnetism.orgplatform (Koymans et al., 2016) was visited and using the tap “Common

Mean True Directions” the reversal test was carried out, which uses theMonte Carlo simulation. It yields a “Class B” result, with an observedangle of 6.21° within a critical angle of 7.40°.

These positive tests imply that demagnetization was successful indefining the primary magnetization.

6. Discussion

This new paleomagnetic pole supersedes the one obtained by Vizán(1998), which was derived from the average of previous data byValencio and Vilas (1970) and Vilas (1974), for the Chon Aike Mag-matic Province. The original data was determined without using or-thogonal diagrams and principal component analyses. As previouslyexplained, this previous pole was calculated for rocks of the Chon AikeMagmatic Province, which by definition has a much wider distributionin Patagonia with a wider range of ages (158.4 Ma by Féraud et al.,1999; to 178.4 Ma by Pankhurst et al., 2000) in comparison to the newpole presented here for the Chon Aike Formation (151.5 to 158.4 Ma,see Fig. 1).

In the Andean Cordillera some 70 km west of the sampled ignim-britic plateau of the Chon Aike Formation the El Quemado Complex, ofcomparable age and similar lithologies outcrops (Fig. 1) and has beenpaleomagnetically studied previously (Iglesia Llanos et al., 2003). Both,the Chon Aike Formation and El Quemado Complex paleomagneticpoles are indistinguishable at the 95% confidence level (Fig. 10). The ElQuemado Complex is not believed to have suffered resetting of itsgeochronological age nor has it been affected by local tectonic rotationsdetectable by paleomagnetism.

Besides this new Chon Aike Formation and the existing El QuemadoComplex paleomagnetic poles, there are no recent paleomagnetic datafor cratonic South America around the age of 155 Ma. To check if theDeseado Massif formed an integral part of South America for LateJurassic times, these paleomagnetic poles were compared with youngerand older reliable poles from NE Brazil (Table 2, Fig. 10). The paleo-magnetic poles from Brazil belong to different dike swarms that havebeen well dated using the K-Ar method. This method, while not as ro-bust as more modern methods, was applied to only highly selected freshrocks following detailed petrographic analysis of the samples. Thosesamples containing glass, clay minerals associated with alteration orsecondary minerals were carefully avoided (Mizusaki et al., 2002).These dates are therefore considered reliable. Age dating, geochemicaland paleomagnetic results allowed division into two clear groups ofdikes, one at 186 Ma and one at 135 Ma (Ernesto et al., 2003).

The poles from Patagonia lie between the Brazilian poles, and their

Fig. 7. Stereographic projection of the mean ChRM directions of the 30 sites. Inred the discarded ones outside a 40° interval with respect to the mean direction(see text). (For interpretation of the references to color in this figure legend, thereader is referred to the web version of this article.)

Fig. 8. Stereographic projection of the AMS results(K1 > K2 > K3) of two different ignimbrites (sites14 & 22) showing similar magnetic fabrics but cor-responding to very different ChRM directions (seeTable 1). This essentially flat flying fabric is re-presentative of ignimbritic flow but also showing aslight inclination of the fabric contrary to the direc-tion of flow (similar to sedimentary fabrics).

V. Ruiz González et al. Tectonophysics 750 (2019) 45–55

51

confidence intervals overlap each other (Fig. 10). This indicates thatsouthern Patagonia was indeed an integral part of the South Americancontinent during the Late Jurassic. Based on previous analyses, vertical axistectonic rotations in the Deseado Massif proposed by Somoza et al. (2008)must be older than 155Ma. On the other hand, the Chon Aike Formationand El Quemado Complex paleomagnetic poles can be used to construct anAPW path for South America during the Mesozoic (see Table 2).

An APW path for South America was constructed (Fig. 11) for theperiod from 203 to 102 Ma using only paleomagnetic poles obtainedfrom igneous rocks to avoid any suspicion of inclination shallowing but

excluding the Lepá-Osta Arena pole (Vizán, 1998) that may have beenaffected by tectonic issues (Zaffarana and Somoza, 2012). These pa-leomagnetic poles (Table 2) are presented in South American geo-graphic coordinates together with the new Chon Aike Formation pole.There are five well-grouped poles from 201 to 198 Ma (14–17 inTable 2), and two with equivalent ages (Anari-Tapirapua Formationand Bolivar Dykes, 12 and 19 in Table 2) that show slightly differentgeographic locations, perhaps due to unresolved tectonic problems.There is another tight group of paleomagnetic poles from 145 to 129 Ma(3–9 in Table 2); and Marifil Fm., Chon Aike Fm. and El QuemadoComplex paleomagnetic poles forming the group in the middle (10–12in Table 2). The Marifil Formation pole would represent a record of thepoleward drift of South America between these two main paleomag-netic pole groups.

Somoza and Zaffarana (2008) proposed a polar standstill betweenca. 125 to at least 100 Ma. Despite the need for more reliable and well-dated paleomagnetic poles of stable South America for the Mesozoicperiod, it is clear that between ~155 Ma and 102 Ma there is very littlechange in the paleomagnetic pole position.

Indeed, all poles for the past 200 m.y. should closely resemble eachother and internal displacement between different blocks within SouthAmerica may be difficult to discern because South America has essen-tially stayed at the same latitude throughout that period (Torsvik et al.,2009).

Kent and Irving (2010) and Kent et al. (2015) motivated bySteinberger and Torsvik (2008) and Torsvik et al. (2012) have sug-gested a very large and rapid phase of TPW between 200 and 140 Ma. ATPW event, by definition, should be detectable in all tectonic plates andwould involve a displacement about an Euler pole located close to theequatorial plane (Marcano et al., 1999). Recently, Fu and Kent (2018)performed an analysis on core samples of the Pacific ocean basin andsuggested that large amplitude TPW occurred as recently as the LateJurassic, however further paleomagnetic experiments on both

Fig. 9. Virtual geomagnetic poles (blue dots) calculated with the mean site directions. The paleomagnetic pole (with its 95% confidence cone, A95 = 8.6°) is shown inpurple. The circumference of the broken red line is the cut-off angle of 40° used in an iterative method to obtain the average pole (see text). (For interpretation of thereferences to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 10. Chon Aike Formation (purple) and El Quemado Complex paleomag-netic poles (deep purple) between the poles obtained from the NE Brazil dikes(Ernesto et al., 1999) (see Table 2). (For interpretation of the references to colorin this figure legend, the reader is referred to the web version of this article.)

V. Ruiz González et al. Tectonophysics 750 (2019) 45–55

52

continental and oceanic crust material during the Late Jurassic to EarlyCretaceous are necessary to test this motion. To evaluate if SouthAmerican paleomagnetic data support this suggestion we compare anAPW path of this continent between ca. 200 Ma and ca. 100 Ma withglobal APW paths for this time span of Torsvik et al. (2012) and Kentand Irving (2010), in South African geographic coordinates, using theEuler rotation parameters proposed by the respective authors (Fig. 12aand b). The South America APW path was constructed using a runningmean with 20 m.y. windows (Table 3). The South American mean polescompare closely with the global mean path of Torsvik et al. (2012)(Fig. 12a). There are some differences between both APW paths but,essentially, the South America path shows a clear counter-clockwiseloop as in the global APW path of Torsvik et al. (2012). In contrast, theSouth American mean poles contrast sharply with the global APW pathof Kent and Irving (2010) differing by some 25° between 160 and140 Ma (Fig. 12b). The track between 180 and 140 Ma of the runningmean APW path for South America does not support the huge shiftsuggested by Kent and Irving (2010), Kent et al. (2015) and Fu and Kent(2018).

7. Conclusions

A reliable paleomagnetic pole which indicates that southernPatagonia was an integral part of South America during the LateJurassic is presented.

Vertical axis tectonic rotations in the Deseado Massif proposed bySomoza et al. (2008) during Gondwana break-up must be older than155 Ma.

Currently, reliable South American paleomagnetic poles do not

Table 2Cretaceous and Jurassic poles for South America (in South American coordinates).

Age (Ma) Code Rock unit Lat (°S) Lon (°E) A95 (deg) References

102 1 Santo Agostinho Lavas −87.6 315 4.5 Schult and Guerreiro (1980)121 2 Florianopolis Dykes −89.1 3.3 2.6 Raposo et al. (1998)129 3 Paraguay alkaline province −85.4 62.3 3.1 Ernesto et al. (1999)130 4 Sierras Pampeanas in Córdoba −86 75.9 3.3 Geuna and Vizán (1998)132 5 Posadas Formation −87.6 150.8 3.5 Mena et al. (2006)132 6 Central Paraná LIP −84.1 64.4 2.3 Ernesto et al. (1999)132 7 Northern Paraná LIP −83.3 71.4 2.4 Ernesto et al. (1999)133 8 Southern Paraná LIP −84 106.2 1.5 Ernesto et al. (1999)125-145 9 NE Brazil magmatism-1 −85.2 97.6 1.8 Ernesto et al. (2003)155 10 Chon Aike Formation −84.3 191.3 8.6 This paper157 11 El Quemado Complex −81 172 7.6 Iglesia Llanos et al. (2003)183 12 Marifil Formation −83 138 11 Iglesia Llanos et al. (2003)197 13 Anari-Tapirapua Formation −65.5 250.3 3.6 Montes-Lauar et al. (1995)175–198 14 NE Brazil magmatism-2 −78.1 223.9 5.2 Ernesto et al. (2003)198 15 French Guyana dikes −81.2 235.1 4 Nomade et al. (2000)199 16 Roraima dikes −80.1 235.1 6.6 Marzoli et al. (1999)200 17 Cacipore dikes −79.8 208.6 5.2 Ernesto et al. (2003)201 18 South America CAMP mean −75.9 237.6 6.8 Kent and Irving (2010)203 19 Bolivar Dykes −66.9 245.6 4.9 MacDonald and Opdyke (1974)

Fig. 11. Paleomagnetic poles of igneous rock from South America between 203and 102 Ma (see Table 2). The new Chon Aike pole of this paper is marked inpurple. (For interpretation of the references to color in this figure legend, thereader is referred to the web version of this article.)

Fig. 12. Running Mean APW paths of South America (see Table 3) rotated toSouth Africa frame (orange) and compared with previous Global Running MeanAPW paths (gray) from: a) Torsvik et al. (2012); b) Kent and Irving (2010).

V. Ruiz González et al. Tectonophysics 750 (2019) 45–55

53

show a significant amount of APW in Jurassic – Cretaceous times andtherefore do not support a TPW event at this time.

Acknowledgments

We dedicate this paper in the loving memory of our friend andteacher Rubén Somoza, sit tibi terra levis. We thank CONICET (ConsejoNacional de Investigaciones Científicas de Argentina) and UBA(Universidad de Buenos Aires) for the financial support (PIP11220120100200CO and UBACyT 20020150100069BA). Correctionsand helpful comments by the editor and reviewers were truly appre-ciated. We also thank Trond Torsvik for his suggestions for the im-provement of the manuscript, and, special thanks to Conall Mac Niocaillfor helping us to achieve a much-improved manuscript with his sug-gestions and his help in English.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.tecto.2018.10.028.

References

Deenen, M.H.L., Langereis, C.G., van Hisbergen, D.J.J., Biggin, A.J., 2011. Geomagneticsecular variation and the statistics of palaeomagnetic directions. Geophys. J. Int. 186,509–520. https://doi.org/10.1111/j.1365-246X.2011.05050.x.

Echeveste, H., Fernández, R., Llambías, E., Tessone, M., Schalamuk, I., Bellieni, G.,Piccirillo, E., De Min, A., 1999. Ignimbritas tardías de alto grado en la FormaciónChon Aike (Jurásico) Macizo del Deseado, Santa Cruz. XIV Congreso GeológicoArgentino, Actas 2, 182–185, Salta, Argentina.

Ernesto, M., Raposo, M.I.B., Marques, L.S., Renne, P.R., Diogo, L.A., de Min, A., 1999.Paleomagnetism, geochemistry, and 40Ar/39Ar dating of the North-eastern ParanáMagmatic Province: tectonic implications. J. Geodyn. 28, 321–340. https://doi.org/10.1016/S0264-3707(99)00013-7.

Ernesto, M., Bellieni, G., Piccirillo, E.M., Marques, L.S., de Min, A., Pacca, I.G., Martins,G., Macedo, J.W.P., 2003. Paleomagnetic and geochemical constraints on the timingand duration of the CAMP activity in northeastern Brazil. In: Hames (Ed.), TheCentral Atlantic Magmatic Province: Insights From Fragments of Pangea. AGUGeophysical Monograph 136. pp. 129–149. https://doi.org/10.1029/136GM07.

Féraud, G., Alric, V., Fornari, M., Bertrand, C., Haller, M., 1999. 40Ar-39Ar dating of theJurassic volcanic province of Patagonia: migrating magmatism related to Gondwanabreak-up and subduction. Earth Planet. Sci. Lett. 172, 83–96. https://doi.org/10.1016/S0012-821X(99)00190-9.

Fisher, R.A., 1953. Dispersion on a sphere. Proceedings Royal Society A217, 295–305.https://doi.org/10.1098/rspa.1953.0064.

Fu, R.R., Kent, D.V., 2018. Anomalous Late Jurassic motion of the Pacific Plate withimplications for true polar wander. Earth Planet. Sci. Lett. 490, 20–30. https://doi.org/10.1016/j.epsl.2018.02.034.

Geuna, S.E., Vizán, H., 1998. New Early Cretaceous palaeomagnetic pole from CórdobaProvince (Argentina): revision of previous studies and implications for the SouthAmerican database. Geophys. J. Int. 135, 1085–1100. https://doi.org/10.1046/j.1365-246X.1998.00688.x.

Goldreich, P., Toomre, A., 1969. Some remarks on polar wandering. J. Geophys. Res. 74,2555–2567. https://doi.org/10.1029/JB074i010p02555.

Gordon, R.G., Livermore, R.A., 1987. Apparent polar wander of the mean-lithospherereference frame. Geophys. J. R. Astron. Soc. 91, 1049–1057. https://doi.org/10.1111/j.1365-246X.1987.tb01679.x.

von Gosen, W., Loske, W., 2004. Tectonic history of Calcatapul formation, ChubutProvince, Argentina, and the “Gastre fault system”. J. S. Am. Earth Sci. 18, 73–88.https://doi.org/10.1016/j.jsames.2004.08.007.

Iglesia Llanos, M.P., Lanza, R., Riccardi, A.C., Geuna, S.E., Laurenzi, M.A., Ruffini, R.,2003. Palaeomagnetic study of the El Quemado complex and Marifil formation,Patagonian Jurassic igneous province, Argentina. Geophys. J. Int. 154, 599–617.https://doi.org/10.1046/j.1365-246X.2003.01923.x.

Kay, S., Ramos, V., Mpodozis, C., Sruoga, P., 1989. Late Paleozoic to Jurassic silicicmagmatism at the Gondwana margin: Analogy to the Middle Proterozoic in NorthAmerica? Geology 17, 324–328. https://doi.org/10.1130/0091-7613(1989)017<0324:LPTJSM>2.3.CO;2.

Kent, D.V., Irving, E., 2010. Influence of inclination error in sedimentary rocks on theTriassic and Jurassic apparent polar wander path for North America and implicationsfor Cordilleran tectonics. J. Geophys. Res. 115, B10103. https://doi.org/10.1029/2009JB007205.

Kent, D.V., Kjarsgaard, B.A., Gee, J.S., Muttoni, G., Heaman, L.M., 2015. Tracking theLate Jurassic apparent (or true) polar shift in U-Pb-dated kimberlites from cratonicNorth America (Superior Province of Canada). Geochem. Geophys. Geosyst. 16,983–994. https://doi.org/10.1002/2015GC005734.

Kirschvink, J.L., 1980. The least-squares line and plane and the analysis of palaeomag-netic data. Geophys. J. R. Astron. Soc. 62, 699–718. https://doi.org/10.1111/j.1365-246X.1980.tb02601.x.

Koymans, M.R., Langereis, C.G., Pastor-Galan, D., van Hinsbergen, D.J.J., 2016.Paleomagnetism.org: an online multi-platform open source environment for paleo-magnetic data analysis. Comput. Geosci. 93, 127–137. https://doi.org/10.1016/j.cageo.2016.05.007.

Lanza, R., Zanella, E., 1993. Palaeomagnetism of the Ferrar dolerite in the northernPrince Albert Mountains (Victoria Land, Antarctica). Geophys. J. Int. 114, 501–511.

Magnetic Prospecting. In: Lanza, R., Meloni, A. (Eds.), The Earth’s Magnetism: AnIntroduction for Geologists. Springer Berlin Heidelberg, Berlin, Heidelberg, pp.99–126. https://doi.org/10.1007/978-3-540-27980-8_3.

MacDonald, W.D., Opdyke, N.D., 1974. Triassic paleomagnetism of northern SouthAmerica. Am. Assoc. Pet. Geol. Bull. 58, 208–215.

Marcano, M.C., van der Voo, R., Mac Niocaill, C., 1999. True polarwander during thePermo-Triassic. Geodynamics 28, 75–95. https://doi.org/10.1016/S0264-3707(98)00026-X.

Marzoli, A., Renne, P.R., Piccirillo, E.M., Ernesto, M., Gellieni, G., De Min, A., 1999.Extensive 200-million-year-old continental flood basalts of the Central AtlanticMagmatic Province. Science 284, 616–618. https://doi.org/10.1126/science.284.5414.616.

McFadden, P.L., McElhinny, M.W., 1990. Classification of the reversal test in palaeo-magnetism. Geophys. J. Int. 103, 725–729. https://doi.org/10.1111/j.1365-246X.1990.tb05683.x.

Mena, M., Orgeira, M.J., Lagorio, S.L., 2006. Paleomagnetism, rock-magnetism andgeochemical aspects of early Cretaceous basalts of the Paraná Magmatic Province,Misiones, Argentina. Earth Planets Space 58, 1283–1293. https://doi.org/10.1186/BF03352623.

Mizusaki, A.M.P., Thomaz-Filho, A., Milani, E.J., de Cesero, P., 2002. Mesozoic andCenozoic igneous activity and its tectonic control in northeastern Brazil. J. S. Am.Earth Sci. 15, 183–198. https://doi.org/10.1016/S0895-9811(02)00014-7.

Montes-Lauar, C.R., Pacca, I.G., Melfi, A.J., Kawashita, K., 1995. Late Cretaceous alkalinecomplexes, southeastern Brazil: paleomagnetism and geochronology. Earth Planet.Sci. Lett. 134, 425–440. https://doi.org/10.1016/0012-821X(95)00135-Y.

Moreira, P., Echeveste, H., Fernández, R., Hartmann, L., Santos, J., 2006. Ajuste geo-cronologico de la Formacion Chon Aike y mineralizaciones epitermales asociadasmediante determinaciones isotopicas U-Pb SHRIMP en circones. In: XI CongresoGeologico Chileno, Actas. 2. pp. 311–314.

Nomade, S., Theveniaut, H., Chen, Y., Pouclet, A., Rigollet, C., 2000. Paleomagnetic studyof French Guyana Early Jurassic dolerites: hypothesis of a multistage magmaticevent. Earth Planet. Sci. Lett. 184, 155–168. https://doi.org/10.1016/S0012-821X(00)00305-8.

Pankhurst, R., Riley, T., Fanning, C., Kelley, S., 2000. Episodic Silicic Volcanism inPatagonia and the Antartic Peninsula: chronology of magmatism associated with theBreak-up of Gondwana. J. Petrol. 41 (5), 605–625. https://doi.org/10.1093/petrology/41.5.605.

Panza, J.L., Haller, M., 2002. El volcanismo jurásico. Geología y Recursos Naturales deSanta Cruz. In: Haller, M. (Ed.), XV Congreso Geológico Argentino, Relatorio,Capítulo I-7, pp. 89–101.

Raposo, M.I.B., Ernesto, M., Renne, P.R., 1998. Paleomagnetism and 40Ar/39Ar dating ofthe early Cretaceous Florianópolis dike swarm (Santa Catarina Island), SouthernBrazil. Phys. Earth Planet. Inter. 108, 275–290. https://doi.org/10.1016/S0031-9201(98)00102-2.

Schult, A., Guerreiro, S.D.C., 1980. Paleomagnetism of Upper Cretaceous volcanic rocksfrom Cabo de Sto. Agostinho, Brazil. Earth Planet. Sci. Lett. 50, 311–315. https://doi.org/10.1016/0012-821X(73)90103-9.

Somoza, R., Zaffarana, C.B., 2008. Mid-Cretaceous polar standstill of South America,motion of the Atlantic hotspots and birth of the Andean cordillera. Earth Planet. Sci.Lett. 271, 267–277. https://doi.org/10.1016/j.epsl.2008.04.004.

Somoza, R., Vizán, H., Taylor, G.K., 2008. Tectonic rotations in the Deseado Massif,southern Patagonia, during the breakup of Western Gondwana. Tectonophysics 460,178–185. https://doi.org/10.1016/j.tecto.2008.08.004.

Steinberger, B., Torsvik, T.H., 2008. Absolute plate motions and true polar wander in theabsence of hotspots tracks. Nature 452, 620–623. https://doi.org/10.1038/nature06824.

Tarling, D.H., Iorio, M., D'Argenio, B., 1999. Geomagnetic long-term secular variation inItalian lower Cretaceous shallow-water carbonates. Geophys. J. Int. 137, 713–722.https://doi.org/10.1046/j.1365-246x.1999.00820.x.

Torsvik, T.H., Rousse, S., Labails, C., Smethurst, M.A., 2009. A new scheme for theopening of the South Atlantic Ocean and the dissection of an Aptian salt basin.Geophys. J. Int. 177, 1315–1333. https://doi.org/10.1111/j.1365-246X.2009.

Table 3Running Mean APW path for South America rotated to South Africa Frame inSouth African coordinates.

Age (Ma) South Africa Frame (Kentand Irving, 2010)

South Africa Frame (Torsviket al., 2012)

± 10 Ma N A95 Plat Plon Plat Plon

100 1 – −65.20 67.65 −65.20 67.65110 – −59.93 70.55 −59.93 70.55120 3 3.6 −55.38 78.89 −55.38 78.89130 8 2 −52.85 83.66 −52.85 83.66140 6 2.2 −49.52 85.9 −52.44 85.14150 2 8.8 −58.02 89.25 −61.00 88.17160 2 8.8 −58.02 89.25 −61.00 88.17180 2 29.5 −59.05 86.26 −61.96 84.80190 6 7.5 −62.91 76.74 −65.54 73.88200 6 6.5 −65.92 67.71 −68.22 63.32

V. Ruiz González et al. Tectonophysics 750 (2019) 45–55

54

04137.x.Torsvik, T.H., Van der Voo, R., Predeen, U., Mac Niocaill, C., Steinberger, B., Doubrovine,

P.V., van Hinsbergen, D.J.J., Domeier, M., Gaina, C., Tohver, E., Meert, J.G.,McCausland, P.J.A., Cocks, L.R.M., 2012. Phanerozoic polar wander, paleogeographyand dynamics. Earth Sci. Rev. 114, 325–368. https://doi.org/10.1016/j.earscirev.2012.06.007.

Torsvik, T.H., Doubrovine, P.V., Domeier, M., 2014. Super-IAPD, Ivar GiæverGeomagnetic Laboratory, Research Council of Norway (NFR), Centre for EarthEvolution and Dynamics (CEED) and University of Oslo (UiO). http://www.iggl.no/resources.html#software.

Valencio, D.A., Vilas, J.F.A., 1970. Paleomagnetism of some Middle Jurassic lavas fromSouth-east Argentina. Nature 225, 262–264. https://doi.org/10.1038/225262a0.

Vilas, J.F.A., 1974. Paleomagnetism of some igneous rocks of the Middle Jurassic Chon-Aike Formation from Estancia La Reconquista, Province of Santa Cruz, Argentina.Geophys. J. R. Astron. Soc. 39, 511–522. https://doi.org/10.1111/j.1365-246X.1974.tb05469.x.

Vizán, H., 1998. Paleomagnetism of the Lower Jurassic Lepa and Osta Arena formations,

Argentine Patagonia. J. S. Am. Earth Sci. 11, 333–350. https://doi.org/10.1016/S0895-9811(98)00018-2.

Vizán, H., Van Zele, M.A., 2001. Analysis of the early Jurassic geomagnetic data recordedat the Bregia Gorge (Ticino, Switzerland). Phys. Earth Planet. Inter. 125, 19–29.https://doi.org/10.1016/S0031-9201(01)00207-2.

Wilson, R.L., Dagley, P., McCormack, A.G., 1972. Palaeomagnetic evidence about thesource of the geomagnetic field. Geophys. J. R. Astron. Soc. 28, 213–224. https://doi.org/10.1111/j.1365-246X.1972.tb06124.x.

Zaffarana, C.B., López de Luchi, M.G., Somoza, R., Mercader, R., Giacosa, R., Martino,R.D., 2010. Anisotropy of magnetic susceptibility study in two classical localities ofthe Gastre Fault System, central Patagonia. J. S. Am. Earth Sci. 30, 151–166. https://doi.org/10.1016/j.jsames.2010.10.003.

Zaffarana, C.B., Somoza, R., 2012. Palaeomagnetism and 40Ar/39Ar dating from LowerJurassic rocks in Gastre, central Patagonia: further data to explore tectonomagmaticevents associated with the break-up of Gondwana. J. Geol. Soc. 169, 371–379.https://doi.org/10.1144/0016-76492011-089.

V. Ruiz González et al. Tectonophysics 750 (2019) 45–55

55